Group Selection?

[CDarwin]

Some of you have probably heard about this. E.O. Wilson, one of the great reductionists, has abandoned his gene-based notions of kin selection and resurrected group selection theory to explain altruism and the evolution of insect colonies. Essentially, the notion holds that while individuals within a group may suffer relative to other members of the same group in being altruistic, the group itself does better than other groups without altruistic members. Thus the altruistic groups are selected for.

 

Here is a recent article on it in the New Scientist: http://www.newscientist.com/channel/life/mg19726383.900-altruism-is-no-family-matter.html

 

Here's Dawkins' criticism in the same issue: http://www.newscientist.com/channel/opinion/mg19726385.700-comment-the-group-delusion.html

 

You might only be able to read the Dawkins comment.

 

For the life of me I could have sworn there was another one published earlier on the topic in New Scinetist by Wilson himself. I can't seem to find it, though. Bothersome.

 

So what 'dya think? Is there anything to this?

[lucaspa]

I think both have problems. Dawkins is making his usual confusion between "unit of selection" and "unit selected for" when he says "What matters is gene selection." What matters is the individual. That is the unit of selection. Within the individual you have particular alleles that are selected for. But it is the individual as a whole that does better or worse under selection.

 

I couldn't read all of Wilson's article but from your description it sounds like Wilson is saying, in hive insects, that the colony is an individual. And that seems to me to be defensible: after all, half the alleles in the colony come from the queen. The other half come from a drone, but there are a limited number of drones that mate with the queen.

 

So, Wilson can argue that the colony is the unit of selection as an individual. Now, whether he is doing that is another matter. In a hive insect, there is confusion about what is an individual. As we normally look on things, a worker bee is an individual. But it is genetically identical to the sibling workers and the worker lives or dies with the colony. You never see worker bees surviving outside a colony. In that regard, the workers are more akin to the cells in a body instead of individuals.

 

So colonies might compete for resources instead of individual worker bees or queens. In that case, "altruism" would be more alike cooperation between the cells of a multicellular organism. The problem seems to be one of semantics. When Wilson used the term "group" instead of explicitly equating the hive to an individual in terms of natural selection, he allowed misunderstanding to arise. So I expect we are going to see a lot of heat and very little light for a while.

 

But, if we step back objectively and look at what is really being said and realize that Wilson's "group" in terms of hive insects is really an individual in terms of selection, then it looks pretty conventional.

 

Of course, Wilson is talking only about hive insects at the moment. Can he apply it to other groups where the individuality is much more pronounced, i.e, wolves, dolphins, and humans? I think those groups are going to still be explained by "kin selection" and not "group selection".

[Mr Skeptic]

Some of you have probably heard about this. E.O. Wilson, one of the great reductionists, has abandoned his gene-based notions of kin selection and resurrected group selection theory to explain altruism and the evolution of insect colonies. Essentially, the notion holds that while individuals within a group may suffer relative to other members of the same group in being altruistic, the group itself does better than other groups without altruistic members. Thus the altruistic groups are selected for.

 

Well, if we are not talking about insect colonies, I would say that individuals in a group with altruistic individuals do better than individuals in a group without altruistic individuals. Also, the individuals in a group will do better the more altruistic individuals there are. This, of course, applies equally well to the individuals who are not altruistic, who will benefit from the others, but will not risk themselves for the others' benefit; they will benefit even more than the altruistic individuals.

 

So we will have a conflict between group benefit and individual benefit, which seems to me like it could have unexpected consequences. For example, both the altruistic and the greedy individual would do better to surround himself with altruistic individuals, focusing his competition against greedy individuals. Us social mammals have created punishments against others who misbehave (more severe punishments for causing harm, but also some for not helping when expected to), and have developed a "tit for tat" system where we reward those who help us (this justice system also has some altruistic attributes). This forces even the greedy individuals to play nice, which also benefits the group. However, it also allows the greed genes to live on, expressed only when it will not be discovered or punished. Though the group would be better off without these greedy individuals, I don't see any way in which they could be eliminated -- except perhaps when the groups are closely related, playing off greed vs kin, which also seems to be the case for social insects.

 

In the end, it is not clear exactly where the selection is made. The selection is of specific genes, not of individuals, groups, or kin. How genes are distributed would have a large impact on how selection proceeds.

[vampares]

Altruism is an inefficient term for a trend.

 

There is tendency to attach purpose to a generalized goal. This is a consequence of logic. That logic can be ingrained to a predisposition. The trivial, genetic or in situ, reasoning is not a consequence of inductive logic as altruism might suppose.

 

Nationalism is not speciesism, per se.

 

----

 

People should not steal from others. Not stealing -- is this altruism? Why don't we steal?

 

People have a tendency to take things into their own world (sustenance). Perhaps this conflicts with another's world. Given the scenario of ones own world invaded by another, there are two contradictions to stealing:

 

(a) social philosophy -- minor revelation, an informal contract -- prevents invasion, so it is made part of the belief system -- insurance by love of life or by confluence

(b) there is empathy for the world of another, and, possibly, there is an empathy transposed, or a fondness of a teaching or some set of rules

© there is a negative consequence (that is the reaction that others have) to stealing, being stolen from

 

"b" is where altruism finds its root. Possession is a result of "c". Ownership is a human construct of "a" however.

 

----

 

Ants have colonies. In these colonies individuals work for the survival of the colony. Worker ants are incapable of sexual reproduction. Does this genetically enslave them to work for the betterment of the Queen, their only means of genetic preservation?

 

Probably not. Sterility more likely results from the unnecessary disadvantages of schlepping an entire reproductive system around.

 

It is, however, a colony which survives. Rather than selecting individuals in a colony -- this gene pool is considerably smaller. Successful colonies will be more able to share their genes with other colonies. Genetic variation is not as frequent -- but the traits are tested at the colony level, so evolves the ant colony.

 

(On ant theories: there is no outward projection by which one colony f's another; the notion would be furthered by the appearance of a truly dedicated apparatus.)

 

----

 

Ants do not switch colonies or work for other species. Bees sort of "give" us honey. Cows "give" us milk.

 

Domesticated animals are not an anomaly. Domesticated humans would not be either.

 

So are we a product of domestication? If so, how much domestication can take place before altruism is irrelevant or we throw the whole theory (all on it's own) out of kilter?

 

----

 

Faith without works is dead. If I volunteer my time to assist disabled persons, or I donate a large some of my wealth to the local university -- what are the chances my genetic kin will benefit from this? In such a way that generosity, as trait, is bettered.

 

People do not like to pay taxes. Bill Clinton had an idea of voluntary tax which the good citzens of Arkansas (or anyone for that matter -- all money is good money in the state of Arkansas, I don't care who you are) could pay, as a contribution. OK so it was lightly received, but if there were no mandatory taxes for Arkansans: what is the likelyhood they would pay? Perhaps stricken with guilt, or enlightened by the financial Renaissance, Arkansans would see their state, not as a municipal burden but an Institution of worthy of their pride and a projection of their will.

 

----

 

Shut down the puppy mills!

 

C.S. Lewis is quick to tie his fellow human to his love and kindness. After all, why does another man have to suffer every time another makes a gain. Only one can wear his ring; do we all suffer?

 

Religion is greater than you and I. So, by and by, we are given to laugh and smile. No scowls or unhappy faces. For if aggression were not a tired tribulation it would swallow us all, but today, altruism leads the way.

[lucaspa]

Well, if we are not talking about insect colonies, I would say that individuals in a group with altruistic individuals do better than individuals in a group without altruistic individuals. Also, the individuals in a group will do better the more altruistic individuals there are. This, of course, applies equally well to the individuals who are not altruistic, who will benefit from the others, but will not risk themselves for the others' benefit; they will benefit even more than the altruistic individuals.

 

1. That last sentence shows why the situation you propose is not stable: the selfish individuals do better -- in terms of natural selection and reproduction -- than the altruistic individuals.

 

2. Of course, you don't demonstrate how a group with altruistic individuals will do better in a reproductive sense. What genes are being passed on if the altruistic individuals sacrifice themselves for the group? Some of those altruistic individuals will die before they have kids and more will die before they can have the maximum number of kids. Therefore the altruistic alleles will be removed from the population. The selfish individuals will, of course, not sacrifice themselves and they will be the ones with comparative reproductive success.

 

and have developed a "tit for tat" system where we reward those who help us (this justice system also has some altruistic attributes). This forces even the greedy individuals to play nice, which also benefits the group. However, it also allows the greed genes to live on, expressed only when it will not be discovered or punished. Though the group would be better off without these greedy individuals, I don't see any way in which they could be eliminated -- except perhaps when the groups are closely related, playing off greed vs kin, which also seems to be the case for social insects.

 

Work in evolutionary psychology indicates that humans have a genetic module for detecting cheating. This is a way of rejecting the greedy individuals and would eliminate them: greedy individuals would be rejected as potential mates. However, this doesn't increase altruism, but just fairness and cooperation.

 

In the end, it is not clear exactly where the selection is made. The selection is of specific genes, not of individuals, groups, or kin.

 

The selection is of individuals. That is very clear. Genes are selected for. Mayr puts it this way:

 

"Much confusion about this problem can be avoided by considering two separate aspects of the question: 'selection of' and 'selection for'. Let us illustrate this with the sickle cell gene. For the question 'selection of' the answer is the individual who either does or does not carry the sickle cell gene. In a malalrial region the answer to 'selection for' is the sickle cell gene, owing to the protection it gives to its heterogenous carriers. When one makes the distinction between the two questions, it becomes quite clear that a gene as such can never be the object of selection. It is only part of a geneotype, whereas the phenotypes of the individual as a whole (based upon the genotype) is the actual object of selection (Mayr 1997). ...

"The reductionist [Dawkins'] thesis that the gene is the object of selection is also invalid for another reason. It is based on the assumption that each gene acts independently of all other genes when making its contribution of genes to the properties of the phenotype. If this were true, the total contribution of genes to the making of the phenotype would be accounted for by the addition of the action of all individual genes. This assumption is referred to as teh 'additive gene action' assumption. Indeed, some genes, perhaps even many genes, seem to act in such a direct and independent manner. If you are a male with the hemophiliac gene, you will be a bleeder. Many other genes, however, interact with each other. Gene B may enhance or reduce the effects of gene A. Or else the effects of gene A will not occur unless gene B is also present. Such interactions among genes are called epistatic interactions." Ernst Mayr What Evolution Is, pgs 126-127

 

Altruism is an inefficient term for a trend.

 

In evolution, "altruism" has a definite meaning. This is another example of not looking at the specific meaning of a word within a discipline, but trying to apply a generalized definition.

 

In evolutionary biology, altruism is "conferral of a benefit on other individuals at an apparent cost to the benefactor" Futuyma, Evolutionary Biology, pg 765

 

It is very apparent that "not stealing" is not altruism.

 

Successful colonies will be more able to share their genes with other colonies.

 

No, they don't. Instead, successful colonies "reproduce" as queens and workers leave to found new colonies.

 

but the traits are tested at the colony level, so evolves the ant colony.

 

That makes the colony the individual. And in evolution, the individual does NOT evolve. Populations evolve. So it would be the population of colonies derived as "offspring" of the colony that would evolve.

 

Faith without works is dead. If I volunteer my time to assist disabled persons, or I donate a large some of my wealth to the local university -- what are the chances my genetic kin will benefit from this? In such a way that generosity, as trait, is bettered.

 

Humans are different. Our behavior is not strictly programmed by our genes as insect behavior is. In fact, our behavior is only very loosely tied to our genes. That's why it is invalid to try to extrapolate insect altruism to human altruism. Because humans can manipulate ideas, we can extend the concept of our "genome" to people who are completely unrelated to us. Thus we can engage in behavior that benefits people we are not related to at apparent cost to ourselves.

[PhDP]

CDarwin,

 

I'm not an expert on multilevel selection, but it's clear that evolutionary biologists are now open to the notion of group selection. Martin Nowak and Sean H. Rice are actively working on this.

 

T. Ryan Gregory also wrote about group/species selection, with words and no maths. oh I know, it's blasphemy ! but he's doing some interesting research and his articles are easier to read.

[lucaspa]

CDarwin,

 

I'm not an expert on multilevel selection, but it's clear that evolutionary biologists are now open to the notion of group selection. Martin Nowak and Sean H. Rice are actively working on this.

 

We'll have to see where the work leads, if anywhere. It is going to take a lot of very good data to convince people that "group selection" does not reduce to individual selection.